Constant current (CC) based power distribution is widely used in the submarine power supply grid for its robustness against cable impedance and short circuit faults. An input-series-output-parallel (ISOP) modularized ...
Constant current (CC) based power distribution is widely used in the submarine power supply grid for its robustness against cable impedance and short circuit faults. An input-series-output-parallel (ISOP) modularized CC-to-CV converter is be used to provide constant voltage (CV) for the submarine instruments. In this paper, an imbalance control with stratified voltage is proposed for the modularized CC-to-CV converter by switching modules to adjust the power. The power of each power module is decided by the output voltage realizing auto and seamless module switching. Specially, only one module is regulated to adjust the power, other modules are out of control working either in full power or in standby, improving the efficiency for light power conditions. The modeling and analysis of the modularized CC-to-CV converter is also presented, as well as the proposed the control method. Finally, a prototype is built to verify the proposed method.
This article investigates the asynchronous fault detection (FD) problem for fuzzy systems with event-triggered mechanism (ETM). A new dynamic ETM (DETM) is adopted to further reduce the waste of network resources. Con...
This article investigates the asynchronous fault detection (FD) problem for fuzzy systems with event-triggered mechanism (ETM). A new dynamic ETM (DETM) is adopted to further reduce the waste of network resources. Considering the impact of asynchronous premise variables brought by ETM, a design criterion for fuzzy FD filter (FDF) is derived. A reasonable residual evaluation function is constructed and an appropriate threshold is set. To ensure the error dynamics be asymptotically stable with a prescribed $H_{\infty}$ performance, we construct a new Lyapunov function that contains an internal dynamic variable in the ETM. A sufficient condition satisfying the proposed performance index is derived. Finally, we provide a numerical simulation to verify the effectiveness of the proposed asynchronous FD strategy under dynamic event-triggered (ET) communication.
High precision modeling in industrial systems is difficult and costly. Model-free intelligent control methods, represented by reinforcement learning, have been applied in industrial systems broadly. The hard evaluated...
详细信息
High precision modeling in industrial systems is difficult and costly. Model-free intelligent control methods, represented by reinforcement learning, have been applied in industrial systems broadly. The hard evaluated of production states and the low value density of processing data causes sparse rewards, which lead to an insufficient performance of reinforcement learning. To overcome the difficulty of reinforcement learning in sparse reward scenes, a reinforcement learning method with reward shaping and hybrid exploration is proposed. By perfecting the rewards distribution in the state space of environment, the reward shaping can make the state-value estimation of reinforcement learning more accurate. By improving the rewards distribution in time dimension, the hybrid exploration can make the iteration of reinforcement learning more efficient and more stable. Finally, the effectiveness of the proposed method is verified by simulations.
Since landslide is one of the most universal natural disasters in China, the study of regional landslide susceptibility evaluation is important to protect people's lives and property. This paper analyzes the geosp...
Since landslide is one of the most universal natural disasters in China, the study of regional landslide susceptibility evaluation is important to protect people's lives and property. This paper analyzes the geospatial characteristics of the Zigui-Badong section in the Three Gorges. By Pearson correlation analysis methodselects, nine impact factors of landslide susceptibility are extracted from the aspects of topography and geomorphology, geological environment, and hydrological conditions, used to establish the evaluation index system of landslide susceptibility. On the above data basis, the paper applies a support vector machine (SVM) model and an SVM model for gray wolf optimization (GWO) to the susceptibility evaluation of landslides, and product landslide susceptibility index maps according to the results. The research area is divided into four regions by jenks method on the map: high-risk, medium-risk, low-risk, and very low-risk areas. Applying the accuracy, confusion matrix, and receiver operating characteristic (ROC) curve to evaluate the model, The prediction accuracy of the GWO-SVM model and the SVM model is 88.55 % and 82.82 % respectively, the comparison proves that the GWO-SVM model is much more accurate, which can provide a reference for the study of regional landslide susceptibility.
This paper explores the finite-time synchronization of a class of discrete-time nonlinear singularly perturbed complex networks using a dynamic event-triggered mechanism (DETM). The DETM is designed to optimize packet...
This paper explores the finite-time synchronization of a class of discrete-time nonlinear singularly perturbed complex networks using a dynamic event-triggered mechanism (DETM). The DETM is designed to optimize packet transmission, aiming to conserve network resources. By constructing a Lyapunov function considering singularly perturbed parameters (SPPs) and DETM information, a sufficient condition for the dynamics of synchronization error system to be finite-time stable is given. The parameters of the synchronization controller can be determined by solving a set of matrix inequalities. The effectiveness of the proposed controller is demonstrated through a numerical example.
Due to the significant time lag and under-regulation, predicting the blast furnace gas generation and formulating its scheduling strategy is complex. This paper proposes a blast furnace gas generation prediction metho...
详细信息
Due to the significant time lag and under-regulation, predicting the blast furnace gas generation and formulating its scheduling strategy is complex. This paper proposes a blast furnace gas generation prediction method based on time series feature extraction and designs a blast furnace gas scheduling strategy based on the prediction results. Firstly, Pearson correlation analysis is used to identify the parameters that have a significant correlation with the blast furnace gas generation, and the selected parameters are decomposed into several intrinsic mode components with different frequency characteristics using the complete ensemble empirical mode decomposition; Then, the principal component analysis method is used to extract the principal components of several intrinsic modal components, and these principal components are employed as the inputs of long short-term memory neural network to predict the blast furnace gas generation; Finally, according to the prediction results designs the scheduling strategy of blast furnace gas. The experiment and contrast experiments are carried out with the industrial field data, and experimental results illustrate that the proposed method is correct and effective.
Effective identification of faults or abnormal conditions can help operators make corrective decisions and plan equipment maintenance. Sequence matching and cluster analysis are important methods to distinguish differ...
Effective identification of faults or abnormal conditions can help operators make corrective decisions and plan equipment maintenance. Sequence matching and cluster analysis are important methods to distinguish different faults. Most existing sequence matching methods mainly focus on alarm event sequences, which reflect the amplitude change characteristics of process data. However, due to the complexity of the equipment and the coupling between variables, alarm event sequences caused by different faults may still assemble each other in a certain extent, which makes it difficult to distinguish faults based on alarms only. To solve this problem, this paper proposes a sequence similarity analysis method combining both alarm and trend events. A qualitative trend representation method is proposed to extract trend changes as trend events. A feature event fusion method is proposed to generate a hybrid sequence to distinguish different fault sequences. The proposed method is evaluated based on data generated by the Tennessee Eastman process model.
Landslide disasters are extremely destructive. Accurate identification of landslides plays an important role in disaster assessment, loss control and post-disaster reconstruction. This paper proposes a semantic segmen...
Landslide disasters are extremely destructive. Accurate identification of landslides plays an important role in disaster assessment, loss control and post-disaster reconstruction. This paper proposes a semantic segmentation landslide identification method based on improved U-Net. The deep convolution neural network and jump connection method is used for end-to-end semantic segmentation to achieve deep feature extraction and fusion of different receptive fields, thus enriching feature information. SENet modules are adopted to enhance the ability of the model to extract important features, so as to further improve the accuracy of model recognition. Extensive experiments show that our improved U-Net achieves better performance than the original algorithm on our landslide datasets. The results of Iou are improved by 4.12% which demonstrates our work is of great significance for the research of landslide area identification. Finally, the model is deployed to the web and applied to the geological hazard intelligent monitoring system to realize the landslide identification task.
The irradiance-power curve is an important basis for examining the operating status of photovoltaic power stations. In the actual operation process, sensor failure, abnormal communication and equipment damage will bri...
详细信息
The irradiance-power curve is an important basis for examining the operating status of photovoltaic power stations. In the actual operation process, sensor failure, abnormal communication and equipment damage will bring a large number of abnormal values to the output data of photovoltaic power plants. It will have a significant impact on a variety of applications based on photovoltaic output data. This paper analyzes the typical outliers on the irradiance-power curve and proposes a photovoltaic output data cleaning method based on fuzzy clustering algorithm and quartile algorithm. By comparing with the quartile method, it is proved that this method can effectively identify abnormal data when there are a large number of outliers in the photovoltaic output data.
A high-reliability constant current to constant voltage power supply system has the advantages of small volume of switching power supply, high power density, high efficiency was proposed. This paper use two controller...
A high-reliability constant current to constant voltage power supply system has the advantages of small volume of switching power supply, high power density, high efficiency was proposed. This paper use two controllers to control the shunt regulator(SR) circuit and single-end flyback converter part, and separate the two parts for small signal modeling and give the parameters to stabilize the closed loop. The state space average modeling idea was used to solve the state equations for the modes of the converters in a switching cycle. In order to ensure the stability of cascade system, this paper collaborative optimization of hardware filter parameters and the appropriate PI parameter design. The experimentals verify the correctness of our theory, and the system has good stability under closed-loop conditions.
暂无评论